Product Safety & Risk Assessment, a Bayesian Network Approach

Product Safety & Risk Assessment, a Bayesian Network Approach.

A causal Bayesian network approach to consumer product safety and risk assessment. Bayesian statistics is a modern branch of statistics based on Bayesian probability, one of the interpretations of probability. In it, probabilities are constantly revised based on new information that has become available.

This independent report was prepared by Queen Mary University London on behalf of the Office for Product Safety and Standards. It outlines the limitations of the risk assessment approach for Safety Gate – formerly RAPEX – the EU’s rapid alert system for dangerous non-food products. An improved methodology based on causal Bayesian networks is proposed.

This is a report for the Office for Product Safety and Standards (OPSS) by Joshua Hunte, Martin Neil and Norman Fenton of Queen Mary University London.

The views expressed in this report are those of the authors and not necessarily those of the Office for Product Safety and Standards or the Department for Business, Energy & Industrial Strategy (nor do they reflect government policy).